A General Parametrization of Correlation Matrix for a Multidimensional Stochastic Volatility Model. Issue 92 (21st November 2017)

• Record Type: Journal Article
• Authors: Sahel, Fouad
• Is Part Of: Wilmott. Volume 2017:Issue 92(2017)
• Page Start: 36
• Page End: 43
• Publication Date: 2017-11-21
• Subjects: correlation matrix -- model risk; Finance -- Periodicals; Financial services industry -- Periodicals; 332
• DOI: 10.1002/wilm.10632
• Languages: English
• ISSNs: 1540-6962

Abstract

Abstract : Recent evolutions in the business of exotic products have rendered the use of stochastic volatility models necessary. Calibration of single stochastic volatility models has already been discussed in several articles. The purpose of this paper is to build a parametrization of the correlation matrix of a multidimensional model with stochastic volatility, given that: the correlation between each asset and its volatility is specified; the correlations between different assets are specified. In the following, we propose a parametrization of the set satisfying the two constraints. This parametrization can be used to calibrate the correlations between single underlying models, and provides a range of attainable prices for standard exotic prices.